Holy Bit Bucket
Was my first response after reading some of the recent news on Tabula. If you Google "Tabula FPGA" you will find a link to the company and a bunch of recent articles. The company appears to be building buzz (hmmm, wonder if they have facebook and twitter accounts) about their technology and future products.
There has been some discussions (comp.arch.fpga) and a bunch of small articles about the relatively new FPGA company. The company is trying to increase current FPGA density by as much as 8x while preserving the cost of current FPGAs. They are essentially using a similar design as current FPGAs but with a twist. They call their technology "Spacetime". "Spacetime" because an internal high clock rate is used to time-division-multiplex a LUT giving the illusion of 8x number of LUTs. In theory they use the same routing resources and number of LUTs as current FPGAs but achieve 8x LUT density. They use a 3D analogy to visualize the increase in logic density. See the Tabula whitepapers for more detailed description of the technology.
The discussion so far on comp.arch.fpga is cautioned excitement. Which was my first response. It is hard not to get excited about the claims but they will have to be commercially successful before the technology is broadly available (chicken and egg).
Many developers are excited about the new technology but also feel they will not have access to the parts any time soon, assuming a premium will be paid for many years before it is widely available, time will tell. Maybe they will partner with some FPGA board designers so that inexpensive development boards will be available. I had the same excitement about Achronix. At the company I was working for we wanted to explore the possibilities of the Achronix FPGAs but the development boards were more than management was willing to pay to test out a new technology. And the development boards were narrowly focused for a particular market. They (management) were content with the current devices so spending more than petty cash wasn't worthwhile.
A general broad customer base might not be the market either of these companies are shooting for (at first). If they are, I think both these companies will need to provide low cost development systems, under $500, so that engineers can convince management to prototype the technology. And so the technology is also available to students and researchers. As one comp.arch.fpga commented it appears Tabula is targeting the high-end networking which is the same as Achronix. Inexpensive development boards might not be an option anytime soon.
Previous post by Christopher Felton:
Next post by Christopher Felton:
Developing FPGA-DSP IP with Python
To post reply to a comment, click on the 'reply' button attached to each comment. To post a new comment (not a reply to a comment) check out the 'Write a Comment' tab at the top of the comments.
Registering will allow you to participate to the forums on ALL the related sites and give you access to all pdf downloads.